(319c) Holistic Understanding of Nanocrystal Stability

At a very small-scale regime (< 10 nm), the properties of materials strongly depend on the size, shape, and surface of the crystals. Consequently, nanocrystals exhibit functionalities that are markedly different from their bulk counterparts. The broad tunability of the properties at the nanoscale combined with an enhanced surface-to-volume ratio makes nanocrystals highly relevant to various applications ranging from optoelectronics to catalysis to medicinal chemistry.¹ However, nanocrystals often exhibit poor stability compared to bulk materials, thus, limiting their capabilities in all the applications mentioned above. Overcoming such a challenge requires a fundamental understanding of nanocrystal stability. In our research, we have worked towards such a goal by: (i) employing novel characterization techniques and developing syntheses that enable studies of the stability of nanocrystals,^2,3 (ii) investigating fundamental mechanisms of growth and stability of nanocrystals,^2,4 and (iii) integrating approaches from data science to accelerate our understanding and obtain a holistic view of nanocrystal stability.

Utilizing new characterization techniques can benefit our knowledge about nanocrystal stability. Taking Pt nanocrystals on high-surface-area carbon (Pt/C) as electrocatalysts as an example, we employed a novel approach to examine the real-time dissolution of Pt from nanocrystals. Specifically, an acid electrolyte is passed through a flow cell. The flow cell outlet is combined with an inductively-coupled plasma mass spectrometry (ICP-MS). The flow cell enables Pt/C nanocrystals to operate in their native environment, and the electrolyte flowing into the ICP-MS provides rich information about the stability of nanocrystals. In particular, we can distinguish at least three degradation mechanisms: (i) ionic dissolution, (ii) nanocluster detachment, and (iii) surface area changes.

Although novel characterization techniques can provide new insight into nanocrystal stability, synthesis limitations can also impact the investigation of nanocrystal stability with more conventional characterization techniques. Taking semiconductor magic-sized nanocrystals (MSNCs) as an example, we developed colloidal synthesis that enabled a more accessible investigation of MSNC growth and stability.^2,3 MSNCs are very small (< 2nm) nanocrystals with exceptional stability that grow in discrete steps jumping from one size to the next directly without forming intermediate crystallites.^2,5 The reason for the exceptional stability of “magic” size remains poorly understood. However, small sizes of MSNCs present significant challenges in their experimental investigation using conventional structural characterization techniques, e.g., they can be challenging to isolate and image via electron microscopy. We overcome these issues by developing a synthetic protocol for CdSe MSNCs using reactive organo-selenium precursors.^2,6 Our synthesis yields MSNCs up to larger length scales (> 2nm) that can be readily isolated and investigated further.

The synthesis of large MSNCs facilitates the investigation of their growth and stability. To determine the source of their stability, we applied a series of conventional optical, (UV-Visible spectroscopy, photoluminescence, photoluminescence excitation) and structural characterization techniques (X-ray diffraction, transmission electron microscopy). Our results suggest that MSNCs are a series of zinc-blende tetrahedral nanocrystals. To investigate their growth and stability, we isolated a particular size of MSNCs and heated them without precursors (conventional ripening). Our observations indicate that the MSNCs grow in the absence of precursors and exhibit size-dependent growth, and stability. These results from growth experiments were then explained by employing an atomistic model based on 3D and 2D-classical nucleation theory. Our model suggests that the exceptional stability of MSNCs arises from their underlying tetrahedral shape and size-dependent growth barriers.²

Utilizing the insights obtained in the model above, we tried to modify the stability and growth trajectory of MSNCs. In particular, we explored three conditions (i) conventional ripening, (ii) ripening with sacrificial species (small MSNCs), and (iii) ripening with continuous addition of monomer.⁴ Our results demonstrate that, even with remarkably distinct initial conditions, the growth trajectory of MSNCs synchronizes. More importantly, our findings indicated two important factors determining this growth synchronization: step and surface energy. By experimentally modulating the surface energy, we demonstrate methods to tune the stability and growth trajectory of MSNCs. The modification of surface energy also helped us to expand the size range of MSNCs further.^2,4

The characterization techniques, syntheses, and growth experiments described above provide different approaches to gathering new insights about nanocrystal stability but are often resource intensive. Consequently, approaches are needed to accelerate our insights about nanocrystal stability. Here, we revert to our first system of Pt/C nanocatalysts. By combining a design-of-experiments-based (DoE) approach with online ICP-MS, we scan a vast parameter space to investigate the role of pre-treatment protocol on the stability of Pt/C nanocatalysts during electrocatalytic oxygen-reduction reaction. In particular, we employ a central composite design that generates a series of pre-treatment protocols for five different factors with five levels each. We then perform our experiments on the online ICP-MS described above. By running very few experiments (2 orders of magnitude < parameter space), our results identifies at least two key parameters (upper potential limit and potential depth) that affect dissolution. These findings help us to hypothesize key physical mechanisms that affect nanocrystal stability. Finally, we propose strategies to integrate such insights from statistical methods into more modern machine-learning-based algorithms (causal model). Thus, by employing novel characterization techniques, developing new syntheses, mechanistic investigation of growth, and integrating tools from data science, me and my colleagues work towards developing holistic understanding of nanocrystal stability.

(1) Kovalenko, M. V; Manna, L.; Cabot, A.; Hens, Z.; Talapin, D. V; Kagan, C. R.; Klimov, V. I.; Rogach, A. L.; Reiss, P.; Milliron, D. J.; Guyot-Sionnnest, P.; Konstantatos, G.; Parak, W. J.; Hyeon, T.; Korgel, B. A.; Murray, C. B.; Heiss, W. Prospects of Nanoscience with Nanocrystals. ACS Nano 2015, 9, 1012–1057.

(2) Mule, A. S.; Mazzotti, S.; Rossinelli, A. A.; Aellen, M.; Prins, P. T.; Van Der Bok, J. C.; Solari, S. F.; Glauser, Y. M.; Kumar, P. V.; Riedinger, A.; Norris, D. J. Unraveling the Growth Mechanism of Magic-Sized Semiconductor Nanocrystals. J. Am. Chem. Soc. 2021, 143, 2037–2048.

(3) Pun, A. B.; Mule, A. S.; Held, J. T.; Norris, D. J. Core/Shell Magic-Sized CdSe Nanocrystals. Nano Lett. 2021, 21, 7651–7658.

(4) Mazzotti, S.; Mule, A. S.; Pun, A. B.; Held, J. T.; Norris, D. J. Growth Synchronization and Size Control in Magic-Sized Semiconductor Nanocrystals. ChemRxiv 2022.

(5) Pun, A. B.; Mazzotti, S.; Mule, A. S.; Norris, D. J. Understanding Discrete Growth in Semiconductor Nanocrystals: Nanoplatelets and Magic-Sized Clusters. Acc. Chem. Res. 2021, 54, 1545–1554.

(6) Riedinger, A.; Mule, A. S.; Knüsel, P. N.; Ott, F. D.; Rossinelli, A. A.; Norris, D. J. Identifying Reactive Organo-Selenium Precursors in the Synthesis of CdSe Nanoplatelets. Chem. Commun. 2018, 54, 11789–11792.